Since radio waves may be considered infrared radiation of long wave, a hot body would be expected to radiate microwave energy thermally. In order to be a good radiator of microwave energy, a body should be a good absorber. An adequate thermal radiator is a “black body.” The amount of radiation emitted in the Millimeter Wave (MMW) range is about 108 times smaller than the amount emitted in the infrared range. Current MMW receivers, however, typically have at least 105 times better noise performance than infrared detectors, and with some temperature contrast, the remaining 103 may be recovered. This makes passive MMW imaging comparable in performance with current infrared systems, making MMW radiometers a popular choice for sensing thermal radiation. MMW radiometers have been used in many different applications such as remote terrestrial and extra-terrestrial sensing, medical diagnostics and defense applications. MMW electromagnetic radiation windows occur at 35 GHz, 94 GHz, 140 GHz and 220 GHz. The choice of frequency depends on specific applications.
Focal plane arrays are used to form images from radiation received by a reflector antenna. Millimeter wave focal plane array radiometers also have been used in many applications to form images based on thermal sensing of radiated microwave energy. The sensitivity of existing radiometer designs, however, has been limited to about 1 deg K., resulting in poor images.
The principle of operation of the radiometric technique is fully described in the literature. The design of a typical radiometer is based on the technique of comparing the level of electromagnetic noise emitted by an unknown source as a target to a reference or stable noise source. This technique and devices were initially proposed by Dicke [R. H. Dicke, “The Measurement of Thermal Radiation at Microwave Frequencies,” The Review of Scientific Instruments, Vol. 17, No. 7, July 1946].
In a Dicke radiometer circuit, the signals from an antenna are sampled and compared with signals from a reference source maintained at a known constant temperature. This overcomes some of the problems of amplifier instability, but in general does not alter effects resulting from imperfect components and thermal gradients.
While other types of radiometric devices have been used with some success, the Dicke (or comparison) type of radiometer has been the most widely used for the study of relatively low level noise-like MMW signals, especially where the noise signals to be examined are often small in comparison to the internally generated noise level within the radiometer receiver. While there are several types of comparison radiometers, one popular type of radiometer for use in the microwave/millimeter wave frequency bands is that in which an incoming signal to be measured as a target and a standard or calibrated reference noise signal are compared. This type of radiometer compares the amplitude of an unknown noise signal coming from the source to be examined with a known amplitude of a noise signal from a calibration source, and has been found useful in measuring with considerable accuracy the effective temperature of an unknown source.
In the Dicke or comparison type radiometer, the receiver input is switched between the antenna and a local reference signal noise generator. The detected and amplified receiver output is coupled to a phase-sensing detector operated in synchronism with the input switching. The output signal from such a radiometer receiver is proportional to the difference between the temperature of the reference signal source and the temperature of the source viewed by the antenna inasmuch as the phase-sensing detector acts to subtract the background or internal noise of the receiver.
A Dicke radiometer uses an RF switch coupled between an antenna and a radiometer receiver, allowing the receiver to alternate between the antenna and a known reference load termination. The receiver output is connected to a synchronous detector that produces an output voltage proportional to a difference between the antenna and the reference temperature. Null balance operation for the Dicke radiometer has been achieved by coupling in noise from a hot noise diode to the antenna port of the RF switch thereby matching the temperature from standard reference loads.
The sensitivity of radiometer measurements are also often limited by random gain fluctuations in the RF front end, low frequency noise (1/f), and bias in the detector circuits. Over the last decades many special techniques, including Dicke switching, have been implemented to reduce measurement errors. Many of these proposals have not yielded a solution that will allow MMW radiometers to be commercially viable. Also, the high cost of MMW RF receivers has limited the number of channels in the radiometer to a low number, resulting in a requirement to scan both azimuth and elevation to create an image.
A challenging aspect of radiometer sensing is the ability to maintain stability over time and temperature. When multiple channel radiometer systems are used, channel-to-channel stability is a challenge. Some sensor characterization over temperature and digital compensation can occur through look-up tables, which have been used extensively to overcome the stability issues. This technique, however, is costly and requires hours of testing per unit. It is also a roadblock to proliferation of radiometer sensors for commercial applications. It would be advantageous to address sensor stability through linearization while eliminating any characterization to result in a more simple circuit design and lower cost.